Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
11" i7 vs i5 Test(s)
- Thread starter Jazwire
- Start date
- Sort by reaction score
Actually, I think we CAN draw conclusions. Even if the i7 is benefitting from the extra RAM/SSD, that's very useful information to know, and further dispels the myth that the higher hardware configurations automatically hurt battery life in all use cases. Evidently, it isn't so cut and dry.
It actually does make sense, if you cast aside assumptions that because a processor is the "upscale" model, it must by definition absolutely use more power, in all cases. The higher L3 cache and enhancements over the i5 might make it capable of performing certain tasks (like sustained video decoding) while using less power overall, whereas the i5 might have to work harder at these tasks, be less efficient, and waste more energy doing them.
Netflix IS pretty practical real-world usage for some people. Your scenario is real-world usage for others too, so that would also be a good test.
Keep in mind that the websites you visit are also going to have a huge impact on what happens. If they're Flash or video heavy, there's going to be a lot more going on than a bunch of static pages.
i7 has more cache and, as you say, is more powerful. It gets more work done per electron "consumed".
Nice test, thanks.
Thanks, someone else bought up the cache issue, I was not thinking about that originally.
My thoughts on this is leading me to believe.
If you are sipping CPU cycles, using very little CPU, the i5 will likely win by a tiny margin in a battery test.
If you are maxing out your CPU 100% the i5 is very likely to win by a larger margin.
But in that 15%-90% CPU range, I am thinking, "possibly" the i7 is actually better for battery life. (Note: this is more of a hypothesis, than a conclusion at this point.)
If thats the case, the next question is, where do you spend most of your time CPU usage wise?
I was wondering if you could convert a video on both with tune4mac or handbrake. I wonder what the time and power difference would be.
Actually, this statement is wrong when seen from a certain point. More cache just means that the probability is higher that the data is already there (cached) and does not need to be processed (again). This saves processing power. You can not generalise this to "electrons" per se.
I, too, find the results very intriguing. That's the reason why I'm so interested in idle performance, because most of it's lifetime the cpu spends idling.
I routinely keep more than 40 tabs open... 41 right nowOnly one browser though
How is that practical? Especially on an 11 or 13 inch screen?
I routinely keep more than 40 tabs open... 41 right now
Is this for work purposes or pleasure??? 40 seems like a lot if you're casually surfing the web.
Here's a thought. Rather than all the bickering over the results, why not have someone else, or several other people, repeat the experiment with the i5 Air and see if they get the same results. All that is required is that someone with an i5 MBA stream the same movie and take note of the battery level at the same intervals as the original poster. If you get the same numbers, the test is legitimate; if not, there is probably a defect in the original poster's MBA.
I think something's wrong with that i5 rig.
If this makes you i7 owners happy about your purchase.. Good.
Wait for Anandtech's i7 vs i5 battery test comparison though. There's just no physical way the i7 can outlast the i5 setup when all things are equal. The i7 will pull more wattage after idle than the i5 will.
If this makes you i7 owners happy about your purchase.. Good.
Wait for Anandtech's i7 vs i5 battery test comparison though. There's just no physical way the i7 can outlast the i5 setup when all things are equal. The i7 will pull more wattage after idle than the i5 will.
Possibly always a chance there is a problem with my wife's i5.
How about a handful of you guys with i5's stream a 90 min movie (pick The Lorax (HD), if we want to keep everything the same), try to use the same settings as I did (stated in the original post).
I'd even be happy to post your results in the 1st thread.
I don't care which unit is better, I own both, I am not trying to "prove" anything, no agenda, if something is wrong with my wife's i5 I would appreciate very much the info myself so I can exchange it.
I do disagree there is no possible for the i7 to outlast the i5, the CPU% for the Safari and SilverLight processes were noticeably higher on the i5.
The i5 was working harder, going into "turbo boost" than the i7 (my theory) which therefore generated more heat and more battery drain. The i7 pulls more wattage at idle and and max cpu (at or near 100%), but in between 15-80% usage, I am having my doubts based on what I've seen so far.
Last edited:
I'm casually surfing the web with 27 lol.
I have a problem where I open something, think to myself "I might need this in the near future" so I'll leave it open until I eventually clean up my tabs.
And the conclusion is?
From my testing them side by side I feel the battery in the i5 lasts noticeably longer and runs cooler and quieter than the i7. I'm keeping the i5 .... I think, I've got until Monday to decide.
I use my 11" Air for web, email, autocad and concept drawing, some video, photo editing and boot into windows via fusion to run some apps. My i7 gets pretty loud and toasty doing that. The i5 not so much.
People seem to be getting mixed results, all along I thought the i5 was the keeper for best battery life and least noise and heat.
Interested to hear the conclusion from the others testing them side by side. My configurations are with 8gb RAM and 512 SSD.
From my testing them side by side I feel the battery in the i5 lasts noticeably longer and runs cooler and quieter than the i7. I'm keeping the i5 .... I think, I've got until Monday to decide.
I use my 11" Air for web, email, autocad and concept drawing, some video, photo editing and boot into windows via fusion to run some apps. My i7 gets pretty loud and toasty doing that. The i5 not so much.
People seem to be getting mixed results, all along I thought the i5 was the keeper for best battery life and least noise and heat.
Interested to hear the conclusion from the others testing them side by side. My configurations are with 8gb RAM and 512 SSD.
Of course I don't *need* it. However, any gain in FPS for an IGP is definitely desirable. I forget that on forums, you have to use words carefully or people pick them apart.
To say "if you're not entirely sure... you dont need it" is just ignorance. It's people who don't know coming to the conclusion they ... don't know.
You want me to elaborate? The i7 in this case is nearly identical except for the L3 cache, minor GPU difference and a small clock speed difference. Let's not forget that the processor is ULT as well: ie, they're not powerhouses.
While I said that most of my work is on battery but I'd like to play some games once in a while doesn't mean I don't care about FPS. However, in a ultra small notebook such as this where performance for portability is sacrificed, beggers CAN be choosers (at least to some degree).
Now if I was playing minesweeper, I'd agree with you...
Last edited:
It's probably this defect and not likely to be one of a hundred other things.
The problem with this is it's impossible to assume that everyone is running the test under the same conditions. It also will inevitably lead to a thread ripe with people's valid results based on 11", 13" or even other notebooks. Also, we will inevitably get tons of people who report either fallacy or with pertinent information left out.
These tests just provide a general insight. We'll never get 100% accurate results as so many factors can vary. On the other hand, we can get close to 100% (the limit of perhaps) but it would take hundreds of posts and then someone with the time to compile the data (if you want to get that complicated over something as minor as this).
----------
Case in point. (I know, his next post oddly has a signature with his specs... not obvious however)
----------
Nice post. Thank you. It's nice to see another user able to compare and report.
----------
Except when the task would require the i5 to run at a higher power state than the i7.
----------
Hey Jaz, for your next test, maybe include some software that can show power draw and possibly even CPU usage in the (don't know what it's called on a mac and I don't want to say taskbar).. menu bar? Your last test may have benefited if you could see what was drawing more power during the stream. And perhaps, when the i5 drew more, then compare the CPU cycles (to see if some other process was consuming the power)... although it would be difficult to watch power draw and not the media content (don't get distracted! the world depends on these results).
----------
Good god! Is something going around consuming our electrons?! What happens when we run out of electrons?
Last edited:
Heat and misc results
I'm running a comparison test for the 11", I5/8/256 and I7/8/256. I was trying to determine which of the machines produced more heat as it makes it more difficult to work if the keyboard is so damn hot.
Both MBAs were freshly started before loading them with the apps in this order.
Brightness for the screen and keyboard were identical.
Apps:
Dropbox running in the background, nothing being downloaded
Safari with 8 tabs of Apple.com
PPT, Word & Excel with a blank doc
Playing 3 movies, 1 on VLC, 2 on QT
Handbrake encoding the same movie
Findings:
Heat: The I7 heated up much faster.
ON THE AVERAGE,
iStat Pro showed the enclosure base 1/2/3 at 40/40/36deg celsius vs 37/37/36 on the I5. I5 was consistently cooler by 3deg c throughout for the enclosure & heatsink B. HD temp was 78 vs 62 (I7/I5). The Airport card, Mem Bank & Controller were largely similar on both machines.
The I7 fans kicked in at 2min and got to 6500rpm within another 2min compared to the I5 kicked in after 6min and took about 15min to reach 6500rpm.
The area around the function keys was burning on the I7 and noticeably less hot on the I5.
Comfort: Running at full blast, the I7 is definitely hotter to the touch than the I5. The palm area flanking the trackpad is hotter on the I7, enough to make me want to not use the cmptr.
Misc performance:
Starting at the same time, the Handbrake encode initially showed a 10% difference in progress (i.e I5 at 5%, I7 at 15%). But when the I5 was at 50% done, the I7 showed 55%, ie the gap got smaller.
The I5 consistently showed that it has 30min more juice than the I7, though I never ran the test that far. Everything starting at the same time, there was a 10% diff in the battery readout (which slowly grew to 15% diff), though the I7 had 5 charge cycles vs 2 cycles on the I5. Not a really exact test but that wasn't the obj of my comparison.
There were 0 page outs despite all the action going on.
Cooling down:
All the apps were closed at the same time, save for one page in Safari.
The I5 fans slowed down first, 1min later the I7 slowed too. Oddly, the I7 dropped sharply to idle fan speed after 2min whereas the I5 took another 3min to do so.
I let both systems idle for 20min and then took these readings.
Enclosure temp were 34/34/33 on the I7, 35/35/34 on the I5.
HD temp was 36/50, heatsink 32/35 (I7/I5).
Even though the insides of the MBA cooled down, the palm area was initially warmer on the I7 than on the I5 and then it reversed, odd huh? I really didn't expect that.
The whole test took 1hr 15 min, final batt remaining is 44%, 5h 46min on the I7 remaining, and 53%, 5hr 43min on the I5 (I5 had 3 fewer charge cycles, prob not calibrated yet).
I'll be using the I5 more over the next few days, I'll post if I have anything to add.
Make of this what you will, hope it'll help some of you come to a decision.
I'm running a comparison test for the 11", I5/8/256 and I7/8/256. I was trying to determine which of the machines produced more heat as it makes it more difficult to work if the keyboard is so damn hot.
Both MBAs were freshly started before loading them with the apps in this order.
Brightness for the screen and keyboard were identical.
Apps:
Dropbox running in the background, nothing being downloaded
Safari with 8 tabs of Apple.com
PPT, Word & Excel with a blank doc
Playing 3 movies, 1 on VLC, 2 on QT
Handbrake encoding the same movie
Findings:
Heat: The I7 heated up much faster.
ON THE AVERAGE,
iStat Pro showed the enclosure base 1/2/3 at 40/40/36deg celsius vs 37/37/36 on the I5. I5 was consistently cooler by 3deg c throughout for the enclosure & heatsink B. HD temp was 78 vs 62 (I7/I5). The Airport card, Mem Bank & Controller were largely similar on both machines.
The I7 fans kicked in at 2min and got to 6500rpm within another 2min compared to the I5 kicked in after 6min and took about 15min to reach 6500rpm.
The area around the function keys was burning on the I7 and noticeably less hot on the I5.
Comfort: Running at full blast, the I7 is definitely hotter to the touch than the I5. The palm area flanking the trackpad is hotter on the I7, enough to make me want to not use the cmptr.
Misc performance:
Starting at the same time, the Handbrake encode initially showed a 10% difference in progress (i.e I5 at 5%, I7 at 15%). But when the I5 was at 50% done, the I7 showed 55%, ie the gap got smaller.
The I5 consistently showed that it has 30min more juice than the I7, though I never ran the test that far. Everything starting at the same time, there was a 10% diff in the battery readout (which slowly grew to 15% diff), though the I7 had 5 charge cycles vs 2 cycles on the I5. Not a really exact test but that wasn't the obj of my comparison.
There were 0 page outs despite all the action going on.
Cooling down:
All the apps were closed at the same time, save for one page in Safari.
The I5 fans slowed down first, 1min later the I7 slowed too. Oddly, the I7 dropped sharply to idle fan speed after 2min whereas the I5 took another 3min to do so.
I let both systems idle for 20min and then took these readings.
Enclosure temp were 34/34/33 on the I7, 35/35/34 on the I5.
HD temp was 36/50, heatsink 32/35 (I7/I5).
Even though the insides of the MBA cooled down, the palm area was initially warmer on the I7 than on the I5 and then it reversed, odd huh? I really didn't expect that.
The whole test took 1hr 15 min, final batt remaining is 44%, 5h 46min on the I7 remaining, and 53%, 5hr 43min on the I5 (I5 had 3 fewer charge cycles, prob not calibrated yet).
I'll be using the I5 more over the next few days, I'll post if I have anything to add.
Make of this what you will, hope it'll help some of you come to a decision.
13 inch - see my sig.
I just can't kill this thing... watched two hd movies and browsed the web all night... 18% left... est. 1:26 remaining.
I'm a software engineer, so both. It's nice for quick reference. Many of my friends also do this.
11" Macbook Air 2012 i7 8gb, but right now I'm outputting to my thunderbolt monitor.